16TH INTERNATIONAL CONFERENCE ON COMPOSITE MATERIALS

Evaluation of Mechanical Properties in Biodegradable Composites Reinforced with a Natural Fiber

Shinji Ogihara*, Akihisa Okada**, Satoshi Kobayashi*** *Department of Mechanical Engineering, Tokyo University of Science ** Graduate student, Department of Mechanical Engineering, Tokyo University of Science *** Faculty of Urban Liberal Arts, Tokyo Metropolitan University Keywords: biodegradable plastics, bamboo fiber, mechanical properties, Polybutylenesuccinate, fiber reinforced composites

Abstract In the present study, we use a bamboo fiber as the reinforcement and polybutylenesuccinate (PBS) as the matrix. We fabricate long fiber unidirectional composites and cross-ply laminate with different fiber volume fractions (10, 20, 30, 40 and 50wt%). We conduct tensile tests on those composites and evaluate the mechanical properties. In addition, we measure the bamboo fiber strength distribution. We discuss experimentally-obtained properties based on the mechanical properties of constituent materials. Then, we obtain following results. Youngs modulus and tensile strength in BF/PBS increase with increasing fiber weight fraction. However, the strain at fracture showed decreasing trend. Youngs modulus in BF/PBS is predictable by the rules of mixture. Composites tensile strength is lower than Curtins prediction of strength which considers distribution of fiber strength.

1. Introduction Plastics have played significant role in the development of the society because of their versatile nature. However, non-biodegradable nature of plastics presents various challenges to environment. Incineration of plastics produces harmful gases. Persistent nature of plastics makes limited land filling sites full. From necessity of eco-friendly materials, biodegradable plastics have been developed. Biodegradable plastics can be produced from recyclable resources, so their materials can be used without destruction of environment. For this reason, these materials have attracted a lot of attention as substitutes of conventional plastics[1]. However, many of them have lower mechanical properties than conventional plastics. Therefore,

there are many studies to improve the properties of these materials for industrial applications. Natural fiber possesses high specific strength and stiffness due to their low density. Therefore, reinforcing biodegradable plastic with a natural fiber is an effective method without a loss of its biodegradability. On such materials, only some combinations of fibers and matrices have been tried and the properties of them are not well understood. In the present study, we use a bamboo fiber as the reinforcement and polybutylenesuccinate (PBS) as the matrix. We fabricate long fiber unidirectional composites and cross-ply laminates ([0/90]s) with different fiber volume fractions (10, 20, 30, 40 and 50wt%). We conduct tensile tests on those composites and evaluate the mechanical properties. In addition, we measure fiber strength distribution[2]. We discuss experimentally-obtained properties based on the mechanical properties of constituent materials. Youngs modulus and tensile strength in unidirectional composite are discussed through the rules of mixture and Curtins composite strength prediction analysis, respectively. Also Youngs modulus and Poissons ratio in cross-ply laminate are discussed through the laminate theory.

2. Experimental method 2.1 Materials and composite fabrication We use bamboo fibers (Ban, Ltd., Tokushima, Japan) as reinforcement and PBS (Showa High Polymers, Ltd., Tokyo, Japan) is used as matrix. We fabricate long fiber unidirectional composites and cross-ply laminate ([0/90]s) using these materials. Hereinafter, those composites are called BF/PBS.

SHINJI OGIHARA, Akihisa Okada and Satoshi Kobayashi

2.2 Composite fabrication We fabricate unidirectional composites and crossply laminates with different fiber weight fractions (10, 20, 30, 40 and 50 wt%), using a hot press (Imoto corporation, Ltd., Kyoto, Japan)[3]. During processing, the materials were kept at 150 and pressed for 10 min. at a pressure of 10MPa. After that, specimens were cooled naturally in the air. Fig.1 shows schematic specimen configuration. Unidirectional composites and cross-ply laminates are in the same configuration. Specimen size was 100mm long, 10mm wide and 1mm thick. 2.3 Tensile test Tensile tests are performed on the composite specimens and the bamboo fibers at a crosshead speed of 0.5mm/min and tensile load is applied until specimen final fracture. After tensile test for fibers, the fracture surface is observed by using an SEM and the cross sectional area is measured.
0 45 90 [0/90]s

where F is the fracture probability at stress , D and D0 are the specimen and reference fiber diameter, respectively, m is the shape parameter and 0 is the scale parameter. By Weibull plotting based on equation (1), we obtained Weibull parameters m and 0 which are listed in Table 1. Fig.2 shows the resulting Weibull plot of bamboo fibers. We used median rank method for cumulative fracture probability F(). Fig.3 shows relation between tensile strength and diameter in bamboo fiber. A large scatter is found in both fiber diameter and strength distribution. Table 1 Mechanical properties in bamboo fiber Number of samples 92 Shape parameter m 3.41 415 Scale parameter 0 (MPa) Average Youngs modulus (GPa) 19.2
4 lnln(1/(1-F()))-2ln(D/Do) 2 0 -2 -4 -6 18.5 y = 3.4076x - 67.826

[0/90] [90/0]

(a) Schematic of specimens

10

19

19.5 ln

20

20.5

Fig.2
100 1 Unit : mm

Weibull plot results in bamboo fiber.

800 Tensile Strength (MPa)

(b) Specimen configuration Fig.1 Composite specimen for tensile test 3. RESULTS AND DISCUSSIONS 3.1 Tensile test on bamboo fibers All bamboo fibers showed brittle fracture behavior in which the load increased proportionally with time and suddenly dropped at the final fracture. To characterize the fiber strength distribution[4], we used Weibull function which can be expressed by the following equation
D 2 m F ( ) = 1 exp D 0 0

700 600 500 400 300 200 100 0 0 0.1 0.2 0.3 0.4 Diameter (mm) 0.5 0.6

(1)

Fig.3

Relation between tensile strength and fiber diameter in bamboo fibers.

2

Evaluation of Mechanical Properties in Biodegradable Composites Reinforced with a Natural Fiber

Stress (MPa)

Table 2 Mechanical properties in the fiber direction Youngs Tensile Longitudinal modulus strength strain at (GPa) (MPa) fracture (%) 10wt% 3.08 43.2 1.118 20wt% 6.15 55.6 0.885 30wt% 10.74 90.7 0.995 40wt% 12.64 78.4 0.772 50wt% 13.54 95.6 0.897

Transverse Strain Longitudinal Strain

70 60 50 40 30 20 10 0 -0.5 0 0.5 Strain (%) 1

10wt% 20wt% 30wt% 40wt% 50wt%

Table 3 Mechanical properties in 45 degree direction Youngs Tensile Longitudinal modulus strength strain at (GPa) (MPa) fracture (%) 10wt% 1.72 13.48 1.05 20wt% 2.1 10.72 0.882 30wt% 2.55 11.03 0.728 40wt% 2.33 7.12 0.357 50wt% 2.78 10.5 0.385

1.5

Fig.4 Stress-strain curves for BF/PBS in the fiber direction

Transverse Strain Longitudinal Strain

14 12

Table 4 Mechanical properties in the transverse direction Youngs Tensile Longitudinal modulus strength strain at (GPa) (MPa) fracture (%) 10wt% 1.06 7.09 0.664 20wt% 1.63 7.38 0.502 30wt% 1.9 6.29 0.385 40wt% 2.4 4.15 0.18 50wt% 2.33 6.16 0.289

Stress (MPa)

10 8 6 4 2 10wt% 20wt% 30wt% 40wt% 50wt% 1

0 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 Strain (%)

Fig.5 Stress-strain curves for BF/PBS in 45 degree direction

Transverse Strain Longitudinal Strain

3.2 Tensile test in fiber direction Fig.4 shows stress-strain curves for BF/PBS unidirectional composite in the fiber direction. Mechanical properties in the fiber direction are listed in Table 2. It is found that Youngs modulus increases with increasing fiber weight fraction but the strain at fracture shows decreasing tendency. Tensile strength increases with increasing fiber weight fraction for 10, 20 and 30wt% but there is small difference among 30 - 50 wt%. Although there is scattering in experimentally-obtained mechanical properties, reinforcing PBS matrix with long bamboo fibers is an effective method in concerning Youngs modulus and tensile strength in the fiber direction.

8 6 4 2 0 -0.2 -0.1 0

Stress (MPa)

10wt% 20wt% 30wt% 40wt% 50wt% 0.1 0.2 0.3 0.4 0.5 0.6 Strain (%)

Fig.6 Stress-strain curves for BF/PBS in the transverse direction

3

SHINJI OGIHARA, Akihisa Okada and Satoshi Kobayashi

3.3 Tensile test in 45 degree and the transverse direction Figs.5 and 6 shows stress-strain curves for BF/PBS unidirectional composites in 45 degree and the transverse directions, respectively. Mechanical properties in 45 degree and transverse directions are listed in Table 3 and 4, respectively. Tensile strength in 45 degree direction is about 10MPa, and that in the transverse direction is about 6MPa. It is found that fiber weight fraction has a small effect on tensile strength. Youngs modulus and the strain at fracture in 45 degree and the transverse directions show similar tendency with those in the fiber direction. Youngs modulus is increasing with increasing with fiber weight fraction and the strain at fracture shows decreasing tendency. 3.4 Tensile test in BF/PBS cross-ply laminates Fig.7 shows stress-strain curves for BF/PBS crossply laminate ([0/90]s). Mechanical properties in BF/PBS cross-ply laminate are listed in Table 5. It is found that Youngs modulus and tensile strength in BF/PBS laminate increases with increasing fiber weight fraction while the strain at fracture shows decreasing tendency. Experimentally-obtained tensile strength is seems to be valid because it is around half of tensile strength of unidirectional composite strength in the fiber direction. 4. ANALYSIS Based on the experimental results, Youngs modulus and strength in unidirectional composites are discussed through the rules of mixture and Curtins composite strength prediction analysis, respectively. Youngs modulus and Poissons ratio in BF/PBS cross-ply laminate are discussed through the laminate theory. We predicted the mechanical properties of unidirectional composites using fiber properties which are shown in Table 1. We predicted the mechanical properties in BF/PBS laminate using experimentally obtained composite properties which are listed in Table 2, 3 and 4. 4.1 Rules of mixture The rules of mixture is known as a simple method in predicting of Youngs modulus of composite materials. We assumed that fibers in composites are arranged in unidirectional and that fiber weight fraction is uniform throughout the composite. Rules of mixture do not consider Poissons ratio. The composite longitudinal and transverse Youngs modulus can be expressed as

Table 5 Mechanical properties in the cross-ply laminate Longitudinal Youngs Tensile strain at modulus strength fracture (%) (GPa) (MPa) 10wt% 23.14 2.344 1.264 20wt% 34.41 4.199 1.051 30wt% 47.96 6.122 0.913 40wt% 53.79 7.406 0.876 50wt% 56.94 8.306 0.811
Transverse Strain Longitudinal Strain

70 60 Stress (MPa) 50 40 30 20 10 0 -0.5 0 0.5 Strain (%) 1

10wt% 20wt% 30wt% 40wt% 50wt%

1.5

Fig.7 Stress-strain curves for cross-ply laminates

E L = E f V f + E m (1 V f )
ET = E f Em E f Vm + EmV f

(2) (3)

Where EL is Youngs modulus in the fiber direction, ET is Youngs modulus in the transverse direction, Ef is fibers Youngs modulus, Em is matrix Youngs modulus, and Vf is fiber volume fraction. Fig.8 shows comparison between prediction by the rules of mixture and the experimental results in (a) the fiber direction and (b) the transverse direction. Although there is some discrepancy between the experimental results and analytical prediction, we believe Youngs modulus in BF/PBS can be roughly predicted by the rule of mixture.
4.2 Curtins composite strength prediction

Curtin proposed a analytical model which predicts composite strength in the fiber direction by considering the statistical nature of the fiber strength. Under the assumption that fiber break arise the other fibers supports the load equally, Curtin derived the composites strength and strain at fracture as[5]
4

Evaluation of Mechanical Properties in Biodegradable Composites Reinforced with a Natural Fiber

20 Young's modulus (GPa) 15 10 5 0 0

Prediction by rule of mixture Experimental results

250 Tensile strength (MPa) 200 150 100 50 0

Prediction by Curtin's analysis Experimental results

10 20 30 40 Fiber weight fraction (wt%)

50

10 20 30 40 50 Fiber weight fraction (wt%)

(a) fiber direction

Prediction by rule of mixture Experimental results

(a) Tensile strength

2.5 Fracture strain (%) 2 1.5 1 0.5 0
Prediction by Curtin's analysis Experimental results

3.5 Young's modulus (GPa) 3 2.5 2 1.5 1 0.5 0 0

10 20 30 40 Fiber weight fraction (wt% )

50

10 20 30 40 50 Fiber weight fraction (wt%)

(b) fiber transverse direction Fig.8 Youngs modulus in BF/PBS. Comparison between prediction by rule of mixture and experimental results for (a) fiber direction and (b) transverse direction

(b) Strain at fracture Fig.9 Comparison between prediction by Curtins analysis and experimental results for (a) Tensile strength and (b) Strain at fracture

uts

2 m +1 m + 1 = f C + (1 f ) y m+2 m+2

(4)

2 m +1 f = Ef m + 2

(5)

where f is fiber volume fraction, y is yield stress of matrix, and Ef is fiber Youngs modulus. In addition, c is fiber maximum stress in fibers fully fragmented. We use experimental results in Table 1 in prediction. Fig.9 shows comparison between prediction by Curtins analysis and the experimental results for (a) tensile strength and (b) strain at fracture. The experimental results in composite strength and the strain at final fracture are lower than the analytical prediction. The discrepancy between the experimental results and the analytical prediction may be due to the assumption made in Curtins model that assumes the equal load sharing. In other words, Curtin assumed that small effect of stress concentration in adjacent fibers to broken fibers by

Fig.10 SEM image of BF/PBS (30wt%) fracture surface debonding around break fiber. In this situation, there is fiber pull-out in fracture surface. Fig.10 shows SEM image of BF/PBS (30wt%) fracture surface. The fracture surface shows relatively flat appearance. This implies following failure
5

SHINJI OGIHARA, Akihisa Okada and Satoshi Kobayashi

mechanism. When one fiber break, it becomes starting point and fracture advances in plane, and causes the composite fracture. In this situation, the effect of stress concentration in adjacent fibers to break fibers should be considered.
4.3 Laminate theory

Youngs modulus and Poissons ratio in BF/PBS laminate are discussed through the laminate theory. In-plane stress resultant per unit width in a laminate is expressed by {N } = [A]{ }0 + [B ]{ }x (6) x where

[A] = [Q ] (z k z k 1 )
N k k =1

N [B ] = 1 Q z k 2 z k 1 2 2 k =1 Assuming that there is no anti-plane deformation, the average stress can be expressed by 1 {N } = 1 [A]{ }0 = (7) x 2H 2H where 2H is the laminate thickness. The inverse relation is { }0 = a * , a * = 2H [A]1 (8) x If we set the x direction the loading direction, Youngs modulus and Poissons ratio in x direction are a * 1 (9) E x = * , xy = 12* a11 a11 The matrix [A] of cross-ply laminates is expressed by

[ ] (
k

2Q12 0 Q11 + Q 22 2Q [A] = 2t Q11 + Q 22 0 (11) 12 0 0 2Q66 where E1 is Youngs modulus in fiber direction, E2 is Youngs modulus in transverse direction, 12 is Poissons ratio and G12 is shear modulus, respectively. Using this equation, we predicted Youngs modulus and Poissons ratio in BF/PBS laminate. Fig.11 shows comparison between prediction by the laminate theory and experimental results for (a) Youngs modulus and (b) Poissons ratio. Prediction of Youngs modulus in BF/PBS laminates shows good agreement with the experimental results. The discrepancy between experimental results and the analytical prediction of Poissons ratio may be because that the reciprocal theorem is not valid in BF/PBS composites.

Young's modulus (GPa)

{}

10 8 6 4 2 0 0

Experimental results Prediction by laminate theory

[ ]{ } [ ]

10 20 30 40 Fiber weight fraction (wt%)

50

(a) Youngs modulus

Experimental results Prediction by laminate theory

[A] = [Q ] (z k z k 1 ) = 2t [Q ]
N k k =1

+Q

[]

90

(10)
Poisson's ratio

0.5 0.4

where

[]

Q11 Q = Q12 0
0 90

Q12 Q22 0 Q12 Q11 0

0 0 Q66 0 0 Q66

0.3 0.2 0.1 0 0 10 20 30 40 50 Fiber weight fraction (wt%)

[Q]

Q 22 = Q12 0

1 12 21 E2 E Q22 = , Q66 = G12 , 21 = 2 12 1 12 21 E1 and [A] becomes

Q11 =

E1

, Q12 =

21 E1 , 1 12 21

(b) Poissons ratio Fig.11 Comparison between prediction by laminate theory and experimental results for

(a) Youngs modulus and (b) Poissons ratio

6

Evaluation of Mechanical Properties in Biodegradable Composites Reinforced with a Natural Fiber

5. CONCLUSION

In the present study, we fabricated biodegradable unidirectional composite and cross-ply laminate which consist of bamboo fibers and PBS, and evaluated their mechanical properties. (1) Youngs modulus and tensile strength in unidirectional composite and cross-ply laminate increase with increasing fiber weight fraction but strain at fracture shows decreasing tendency. (2) Youngs modulus in unidirectional composites is predictable by the rules of mixture. Unidirectional composites tensile strength is lower than Curtins prediction of strength which considers distribution of fiber strength. (3) Youngs modulus in BF/PBS cross-ply laminate is predictable by the laminate theory. However, analytical prediction of Poissons ratio in BF/PBS cross-ply laminate is lower than the experimental results.
REFERENCE 1. H.Tsuji, Biodegradable polymers, Corona Publishing Co., Tokyo (2002), pp.1-15. (in Japanese) 2. K.Takura, Proceedings of the 30th Composites Symposium, Japan Society of Composite Materials (2001), pp.115-116. 3. Y.Kitamura, Proceedings of the 32th Composites Symposium, Japan Society of Composites Materials (2003), pp.29-30 4. K.Suzuki, Materials, 54 (2005), pp.887-894 (in Japanese) 5. G.Ben and T.Ishikawa, Advanced Composites Engineering, Baifukan (2005), pp.13-20. (in Japanese)